-
Notifications
You must be signed in to change notification settings - Fork 0
/
index.html
457 lines (451 loc) · 19.1 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
<!doctype html>
<html>
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0, maximum-scale=1.0, user-scalable=no">
<title>Nanopore sequencing for Mycobacterium tuberculosis drug susceptibility testing and outbreak investigation</title>
<link rel="stylesheet" href="dist/reset.css">
<link rel="stylesheet" href="dist/reveal.css">
<link rel="stylesheet" href="dist/theme/nord.css">
<!-- Theme used for syntax highlighted code -->
<link rel="stylesheet" href="plugin/highlight/monokai.css">
</head>
<body>
<!-- Border line around slides -->
<div class="line top"></div>
<div class="line bottom"></div>
<div class="line left"></div>
<div class="line right"></div>
<!-- Border line around slides -->
<div class="reveal">
<div class="slides">
<section>
<h5>Nanopore sequencing for <em>Mycobacterium tuberculosis</em> drug susceptibility testing and outbreak investigation</h5>
<br>
<p>Michael B. Hall $\in$ Iqbal Group @ EMBL-EBI</p>
<p style="font-family: monospace;font-size: 0.5em"><a href="https://twitter.com/mbhall88">@mbhall88</a> | michael.hall2 [at] unimelb.edu.au</p>
<p><a href="https://mbh.sh/qmrl-slides/">https://mbh.sh/qmrl-slides/</a></p>
</section>
<section>
<p class="fragment fade-in-then-semi-out">Illumina increasingly being used for TB clustering and DST (England, the Netherlands, New York State)</p>
<p class="fragment">Requires large capital outlay and significant testing volume to be cost-effective and ensure clinically appropriate turnaround times</p>
</section>
<section>
<p class="fragment fade-in-then-semi-out">ONT's MinION device offers solution to these two issues</p>
<p class="fragment fade-in-then-semi-out">Error rate has been an obstacle in the past</p>
<p class="fragment">Some early signs that error rate has sufficiently improved for DST and clustering</p>
</section>
<section>
<a href="https://doi.org/10.1128/JCM.00583-20" target="_blank"><img src="imgs/smith2021.png"></a>
<ul>
<li>431 isolates sequenced on both platforms</li>
<li>DST results highly concordant</li>
<li>Clustering lacked detail</li>
<li>In-house scripts and catalogues</li>
</ul>
</section>
<section>
<h2>Aim</h2>
<p>Compare Nanopore to Illumina in its ability to: predict drug resistance and identify putative transmission clusters using SNPs</p>
</section>
<section>
<h2>Acknowledgements</h2>
<div class="row">
<div class="column">
<b>Madagascar</b><br>
<ul>
<li>Marie Sylvianne Rabodoarivelo</li>
<li>Niaina Rakotosamimanana</li>
<li>Simon Grandjean Lapierre</li>
</ul><br><br>
<b>EMBL-EBI</b><br>
Zamin Iqbal
</div>
<div class="column">
<b>South Africa</b><br>
<ul>
<li>Anastasia Koch</li>
<li>Anzaan Dippenaar</li>
<li>Helen Cox</li>
<li>Robin Warren</li>
<li>Melanie Grobbelaar</li>
</ul><br><br>
<b>Oxford</b><br>
<ul>
<li>Tim Walker</li>
<li>Tim Peto</li>
<li>Sophie George</li>
</ul>
</div>
</div>
</section>
<section>
<h2>Data</h2>
<ul>
<li class="fragment fade-in-then-semi-out">208 isolates from Madagascar ($n$=109), South Africa (67), England (32)</li>
<li class="fragment fade-in-then-semi-out">Sequenced from same DNA extraction on Illumina and Nanopore</li>
<li class="fragment">35 Malagasy isolates sequenced on PacBio CCS</li>
</ul>
</section>
<section>
<h2>Quality control</h2>
<ul>
<li class="fragment fade-in-then-semi-out">Nanopore basecalled with Guppy <b>v5.0.16</b></li>
<li class="fragment fade-in-then-semi-out">Remove non-Mtb reads (all data)</li>
<li class="fragment fade-in-then-semi-out">Decontaminated read depth $\ge20$ (Illumina; $n$=1) $\ge30$ (Nanopore; $n$=37) with 6 failing both</li>
<li class="fragment fade-in-then-semi-out">Remove mixed and indeterminate lineage isolates ($n$=12)</li>
<li class="fragment fade-in-then-semi-out">One isolate found to have non-matched sequencing data</li>
<li class="fragment">151 total isolates after QC</li>
</ul>
</section>
<section>
<h2>Variant calling</h2>
<p class="fragment fade-in-then-semi-out">SNP-based requires high-quality SNP calls</p>
<p class="fragment fade-in-then-semi-out">Illumina: <a href="https://github.com/oxfordmmm/CompassCompact">COMPASS</a> (stampy, samtools, and BCFtools)</p>
<span class="fragment fade-in"><span class="fragment highlight-red">Nanopore: Minimap2 and BCFtools (<b>v1.13</b>) with custom filtering</span></span>
<ul>
<li class="fragment fade-in">Quality score $\lt85$</li>
<li class="fragment fade-in">$\lt1$% reads on both strands</li>
<li class="fragment fade-in">Read depth $\lt20$% isolate median</li>
<li class="fragment fade-in">Variant distance bias (VDB) $\lt0.00001$</li>
<li class="fragment fade-in">$\lt90$% reads supporting called allele</li>
</ul>
</section>
<section>
<h2>SNP evaluation</h2>
<p class="fragment fade-in-then-semi-out">Seven isolates with PacBio CCS data over 20x read depth</p>
<p class="fragment fade-in-then-semi-out">Assemble with <a href="https://github.com/fenderglass/Flye" target="_blank">Flye</a> and polish for indels with Illumina</p>
<p class="fragment fade-in-then-semi-out">Evaluate precision and recall with <code><a target="_blank" href="https://github.com/iqbal-lab-org/varifier">varifier</a></code> and
<code><a href="https://github.com/Illumina/hap.py" target="_blank">hap.py</a></code></p>
</section>
<section>
<p>Recall = $\frac{TP}{TP+FN}$ | Precision = $\frac{TP}{TP+FP}$</p>
<img width="600" src="imgs/filter-precision-recall.png">
</section>
<section>
<h2>SNP distance</h2>
<p><code><a href="https://github.com/mbhall88/psdm">psdm</a></code> to generate distance matrix</p>
<img width="550" src="imgs/full.dotplot.png">
</section>
<section>
<h2>SNP distance</h2>
<p>Isolates within (Illumina) distance of 20 SNPs</p>
<img width="550" src="imgs/close.dotplot.png">
</section>
<section>
<h2>Evaluating SNP-based transmission clusters</h2>
<p>Illumina thresholds of interest 5 and 12</p>
<p class="fragment fade-in-then-semi-out">Take Illumina clustering as truth and see how similar Nanopore clustering is</p>
<p class="fragment fade-in-then-semi-out">What Nanopore thresholds to use?</p>
<p class="fragment fade-in">How to define similarity?</p>
</section>
<section>
<h2>Cluster similarity - recall</h2>
<p><b>SACR</b> - sample-averaged cluster recall</p>
<p>$$\frac{\left|C_{n,I}\cap C_{n,N}\right|}{|C_{n,I}|}$$</p>
<p>$C_{n,I}$ - the Illumina ($I$) cluster of isolate $n$</p>
<p>$C_{n,N}$ - the Nanopore ($N$) cluster of isolate $n$</p>
<p>$\cap$ - intersection</p>
<p>$|x|$ - size/length of $x$</p>
</section>
<section>
<h2>Cluster similarity - precision</h2>
<p><b>SACP</b> - sample-averaged cluster precision</p>
<p>$$\frac{\left|C_{n,I}\cap C_{n,N}\right|}{|C_{n,N}|}$$</p>
<p>Same as SACR, but denominator is size of Nanopore cluster</p>
</section>
<section>
<h2>Cluster similarity - excess clustering rate</h2>
<p>SACR and SACP don't identify Nanopore clusters solely composed of Illumina singletons</p>
<p>$$XCR = \frac{|S_I-S_N|}{|S_I|}$$</p>
<p>$S_I$ - Illumina singletons</p>
</section>
<section>
<p>Example: Illumina (left) and Nanopore (right)</p>
<img height="400" src="imgs/illumina-cluster-example.png">
<img height="400" src="imgs/ont-cluster-example.png">
<style type="text/css">
.tg {border-collapse:collapse;border-spacing:0;border-bottom: 1px solid black}
.tg td{border-color:black;border-style:solid;border-width:1px;font-size:22px;
overflow:hidden;padding:10px 5px;word-break:normal;}
.tg th{border-color:black;border-style:solid;border-width:1px;font-size:22px;
font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;}
.tg .tg-0pky{border-color:inherit;text-align:left;vertical-align:top}
@media screen and (max-width: 767px) {.tg {width: auto !important;}.tg col {width: auto !important;}.tg-wrap {overflow-x: auto;-webkit-overflow-scrolling: touch;}}</style>
<div class="tg-wrap"><table class="tg">
<thead>
<tr>
<th class="tg-0pky">sample</th>
<th class="tg-0pky">A</th>
<th class="tg-0pky">B</th>
<th class="tg-0pky">C</th>
<th class="tg-0pky">D</th>
<th class="tg-0pky">E</th>
<th class="tg-0pky">F</th>
<th class="tg-0pky">G</th>
<th class="tg-0pky">H</th>
<th class="tg-0pky">I</th>
<th class="tg-0pky">J</th>
<th class="tg-0pky">avg.</th>
</tr>
</thead>
<tbody>
<tr>
<td class="tg-0pky">recall</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">0.8</td>
<td class="tg-0pky">0.8</td>
<td class="tg-0pky">0.8</td>
<td class="tg-0pky">0.8</td>
<td class="tg-0pky">0.0</td>
<td class="tg-0pky"><b>0.82</b></td>
</tr>
<tr>
<td class="tg-0pky">precision</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">0.66</td>
<td class="tg-0pky">0.66</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">1.0</td>
<td class="tg-0pky">0.0</td>
<td class="tg-0pky"><b>0.83</b></td>
</tr>
</tbody>
</table></div>
<p>20 Illumina singletons - XCR of $2/20=0.1$.</p>
</section>
<section>
<h4>Selecting Nanopore SNP thresholds</h4>
<img width="550" src="imgs/cluster_threshold_sweep.png">
</section>
<section>
<h3>5-SNP clusters</h3>
<img src="imgs/clusterst5.png" height="600">
</section>
<section>
<h3>12-SNP clusters</h3>
<img src="imgs/clusterst12.png" height="600">
</section>
<section>
<h3>Mixed technology</h3>
<img src="imgs/mixed_self_distance.png" height="600">
</section>
<section>
<h3>Mixed technology</h3>
<img src="imgs/mixed.dotplot.png" height="600">
</section>
<section>
<h3>Mixed technology</h3>
<img src="imgs/mixed_simulations.png" height="600">
</section>
<section>
<h2>Drug resistance</h2>
<p class="fragment fade-in-then-semi-out">How do Nanopore WGS-based predictions compare to Illumiuna? (Mykrobe)</p>
<p class="fragment fade-in-then-semi-out">If different, what do phenotypes say?</p>
<p class="fragment fade-in">Primary focus is genotype concordance - thoroughly validated</p>
</section>
<section>
<h3>Mykrobe catalogue</h3>
<ul>
<li class="fragment fade-in-then-semi-out">476 amino acids changes (3,352 nucleotide mutations)</li>
<li class="fragment fade-in-then-semi-out">60 promoter SNPs</li>
<li class="fragment fade-in-then-semi-out">1,904 nucleotide-level SNPs and indels</li>
<li class="fragment fade-in-then-semi-out">All <em>katG</em> and <em>pncA</em> 1/2bp frameshifts (61,258)</li>
<li class="fragment fade-in">Total of 66,574 nucleotide-level resistance mutations</li>
</ul>
</section>
<section>
<h3>Mykrobe genotype concordance</h3>
<p class="fragment fade-in-then-semi-out">One isolate with a 1bp deletion in <em>katG</em></p>
<p class="fragment fade-in">One isolate with three 1bp deletions in <em>katG</em></p>
</section>
<section>
<h3>Mykrobe prediction concordance</h3>
<style type="text/css">
.thick .tg-0pky {
border-bottom-width: 4px;
}
.tg {border-collapse:collapse;border-spacing:0;}
.tg td{border-color:black;border-style:solid;border-width:1px;font-size:20px;
overflow:hidden;padding:10px 5px;word-break:normal;}
.tg th{border-color:black;border-style:solid;border-width:1px;font-size:20px;
font-weight:normal;overflow:hidden;padding:10px 5px;word-break:normal;}
.tg .tg-0lax{text-align:left;vertical-align:top}
.tg .tg-0pky{border-color:inherit;text-align:left;vertical-align:top}
.tg-sort-header::-moz-selection{background:0 0}
.tg-sort-header::selection{background:0 0}.tg-sort-header{cursor:pointer}
.tg-sort-header:after{content:'';float:right;margin-top:7px;border-width:0 5px 5px;border-style:solid;
border-color:#404040 transparent;visibility:hidden}
.tg-sort-header:hover:after{visibility:visible}
.tg-sort-asc:after,.tg-sort-asc:hover:after,.tg-sort-desc:after{visibility:visible;opacity:.4}
.tg-sort-desc:after{border-bottom:none;border-width:5px 5px 0}@media screen and (max-width: 767px) {.tg {width: auto !important;}.tg col {width: auto !important;}.tg-wrap {overflow-x: auto;-webkit-overflow-scrolling: touch;}}</style>
<div class="tg-wrap"><table id="tg-fOdTz" class="tg">
<thead>
<tr>
<th class="tg-0lax">Drug</th>
<th class="tg-0lax">FN(R)</th>
<th class="tg-0lax">FP(S)</th>
<th class="tg-0lax">FNR(95% CI)</th>
<th class="tg-0lax">FPR(95% CI)</th>
<th class="tg-0lax">PPV(95% CI)</th>
<th class="tg-0lax">NPV(95% CI)</th>
</tr>
</thead>
<tbody>
<tr>
<td class="tg-0pky">Isoniazid</td>
<td class="tg-0pky">0(81)</td>
<td class="tg-0pky"><b>1</b>(70)</td>
<td class="tg-0pky">0.0% (0.0-4.5%)</td>
<td class="tg-0pky">1.4% (0.3-7.7%)</td>
<td class="tg-0pky">98.8% (93.4-99.8%)</td>
<td class="tg-0pky">100.0% (94.7-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Rifampicin</td>
<td class="tg-0pky">0(79)</td>
<td class="tg-0pky">0(72)</td>
<td class="tg-0pky">0.0% (0.0-4.6%)</td>
<td class="tg-0pky">0.0% (0.0-5.1%)</td>
<td class="tg-0pky">100.0% (95.4-100.0%)</td>
<td class="tg-0pky">100.0% (94.9-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Ethambutol</td>
<td class="tg-0pky">0(54)</td>
<td class="tg-0pky">0(97)</td>
<td class="tg-0pky">0.0% (0.0-6.6%)</td>
<td class="tg-0pky">0.0% (0.0-3.8%)</td>
<td class="tg-0pky">100.0% (93.4-100.0%)</td>
<td class="tg-0pky">100.0% (96.2-100.0%)</td>
</tr>
<tr class="thick">
<td class="tg-0pky">Pyrazinamide</td>
<td class="tg-0pky">0(30)</td>
<td class="tg-0pky">0(121)</td>
<td class="tg-0pky">0.0% (0.0-11.4%)</td>
<td class="tg-0pky">0.0% (-0.0-3.1%)</td>
<td class="tg-0pky">100.0% (88.6-100.0%)</td>
<td class="tg-0pky">100.0% (96.9-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Streptomycin</td>
<td class="tg-0pky">0(47)</td>
<td class="tg-0pky"><b>1</b>(104)</td>
<td class="tg-0pky">0.0% (0.0-7.6%)</td>
<td class="tg-0pky">1.0% (0.2-5.2%)</td>
<td class="tg-0pky">97.9% (89.1-99.6%)</td>
<td class="tg-0pky">100.0% (96.4-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Amikacin</td>
<td class="tg-0pky">0(13)</td>
<td class="tg-0pky"><b>1</b>(138)</td>
<td class="tg-0pky">0.0% (0.0-22.8%)</td>
<td class="tg-0pky">0.7% (0.1-4.0%)</td>
<td class="tg-0pky">92.9% (68.5-98.7%)</td>
<td class="tg-0pky">100.0% (97.3-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Capreomycin</td>
<td class="tg-0pky">0(13)</td>
<td class="tg-0pky"><b>1</b>(138)</td>
<td class="tg-0pky">0.0% (0.0-22.8%)</td>
<td class="tg-0pky">0.7% (0.1-4.0%)</td>
<td class="tg-0pky">92.9% (68.5-98.7%)</td>
<td class="tg-0pky">100.0% (97.3-100.0%)</td>
</tr>
<tr class="thick">
<td class="tg-0pky">Kanamycin</td>
<td class="tg-0pky">0(14)</td>
<td class="tg-0pky"><b>1</b>(137)</td>
<td class="tg-0pky">0.0% (0.0-21.5%)</td>
<td class="tg-0pky">0.7% (0.1-4.0%)</td>
<td class="tg-0pky">93.3% (70.2-98.8%)</td>
<td class="tg-0pky">100.0% (97.3-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Ciprofloxacin</td>
<td class="tg-0pky">0(16)</td>
<td class="tg-0pky">0(135)</td>
<td class="tg-0pky">0.0% (0.0-19.4%)</td>
<td class="tg-0pky">0.0% (0.0-2.8%)</td>
<td class="tg-0pky">100.0% (80.6-100.0%)</td>
<td class="tg-0pky">100.0% (97.2-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Moxifloxacin</td>
<td class="tg-0pky">0(16)</td>
<td class="tg-0pky">0(135)</td>
<td class="tg-0pky">0.0% (0.0-19.4%)</td>
<td class="tg-0pky">0.0% (0.0-2.8%)</td>
<td class="tg-0pky">100.0% (80.6-100.0%)</td>
<td class="tg-0pky">100.0% (97.2-100.0%)</td>
</tr>
<tr>
<td class="tg-0pky">Ofloxacin</td>
<td class="tg-0pky">0(17)</td>
<td class="tg-0pky">0(134)</td>
<td class="tg-0pky">0.0% (0.0-18.4%)</td>
<td class="tg-0pky">0.0% (0.0-2.8%)</td>
<td class="tg-0pky">100.0% (81.6-100.0%)</td>
<td class="tg-0pky">100.0% (97.2-100.0%)</td>
</tr>
</tbody>
</table></div>
</section>
<section>
<h2>Drug susceptibility testing</h2>
<img src="imgs/available-dst.png" height="600">
</section>
<section>
<h3>Phenotype concordance</h3>
<img src="imgs/phenotype_concordance.png" height="600">
</section>
<section>
<h3>Impact of Nanopore read depth</h3>
<div style="line-height: 0">
<img src="imgs/dst-depth.png" height="300">
<img src="imgs/decumulative-depth.png" height="300">
</div>
</section>
<section>
<h2>Summary</h2>
<ul>
<li class="fragment">Nanopore SNP precision/recall similar to Illumina</li>
<li class="fragment">Nanopore clusters do not miss isolates</li>
<li class="fragment">Mixing technologies is possible</li>
<li class="fragment">Nanopore resistance predictions consistent with Illumina</li>
<li class="fragment">Read depth of 30x does not impact resistance predictions</li>
</ul>
<p><a href="https://mbh.sh/qmrl-slides/">https://mbh.sh/qmrl-slides/</a></p>
<p style="font-family: monospace;font-size: 0.5em"><a href="https://twitter.com/mbhall88">@mbhall88</a> | michael.hall2 [at] unimelb.edu.au</p>
</section>
</div>
</div>
<script src="dist/reveal.js"></script>
<script src="plugin/notes/notes.js"></script>
<script src="plugin/markdown/markdown.js"></script>
<script src="plugin/highlight/highlight.js"></script>
<script src="plugin/math/math.js"></script>
<script>
// More info about initialization & config:
// - https://revealjs.com/initialization/
// - https://revealjs.com/config/
Reveal.initialize({
margin: 0.2,
hash: true,
// Learn about plugins: https://revealjs.com/plugins/
plugins: [ RevealMarkdown, RevealHighlight, RevealNotes, RevealMath.KaTeX ]
});
</script>
</body>
</html>